What will the frontier METR time horizon be on January 1, 2027?
15
Ṁ125Ṁ9872027
0.6%
4-8 Hours
0.6%
8-16 Hours
12%
16-32 Hours
30%
32-64 Hours
57%
The naive extrapolation from the original paper would be around 8 hours. If METR decides to stop publishing the benchmark or there is some other problem, I will mark as N/A.
Update 2026-02-21 (PST) (AI summary of creator comment): If the METR time horizon estimate breaks down or is ambiguous/unknown, the market will resolve to the best point estimate even if the confidence interval is wide.
This question is managed and resolved by Manifold.
Market context
Get
1,000 to start trading!
People are also trading
What will be the METR time horizon doubling time in 2026?
Best METR 50% Time Horizon in 2026
Best AI time horizon by February 2026, per METR?
Will a multi-agent system have its time horizon evaluated by METR before August 2026?
37% chance
Will METR retire the 50% Time Horizon by EOY 2026
40% chance
Will the METR 50% Time Horizon be "ambiguous" at the end of 2026?
62% chance
What will the frontier METR time horizon be on January 1, 2028?
Best AI time horizon by August 2026, per METR?
Grok 4.20 METR 50% time horizon
R2 / V4-Thinking METR 50% time horizon
Sort by:
@DavidHiggs It will be the best point estimate, even if the CI is wide. Should I add a "64-128 hour" bucket? Or "64+ hours"?
People are also trading
Related questions
What will be the METR time horizon doubling time in 2026?
Best METR 50% Time Horizon in 2026
Best AI time horizon by February 2026, per METR?
Will a multi-agent system have its time horizon evaluated by METR before August 2026?
37% chance
Will METR retire the 50% Time Horizon by EOY 2026
40% chance
Will the METR 50% Time Horizon be "ambiguous" at the end of 2026?
62% chance
What will the frontier METR time horizon be on January 1, 2028?
Best AI time horizon by August 2026, per METR?
Grok 4.20 METR 50% time horizon
R2 / V4-Thinking METR 50% time horizon